Converting yyyymmdd double to decimal year
Show older comments
Hi there, I have many variables made from doubles. I would like to plot certain variables against the date which is also made of doubles in the format yyyymmdd. So far i have:
function [out1] = simpleplot(variable,depthvar,mindepth,maxdepth,datevar,startdate,enddate,xlab,ylab);
depth = find(depthvar>mindepth & depthvar<maxdepth & ~isnan(variable));
datedepth=datevar(depth);
variable_depth=variable(depth);
plot(datedepth, variable_depth,'sk--')
xlabel(xlab);
ylabel(ylab);
axis([startdate enddate min(variable_depth) max(variable_depth)]);
end
However because my dates are yyyymmdd not in decimal year the plot has large gaps in it.
How can i convert the date doubles to decimal year within the function and also to have the x axis show date in a meaningful way like yy/mm rather than decimal year.
thank-you
Accepted Answer
More Answers (1)
Star Strider
on 14 Feb 2016
To get MATLAB date numbers from your date data, this works:
date_var = 20160214;
date_number = datenum(sprintf('%.0f', date_var), 'yyyymmdd');
Check = datestr(date_number)
Check =
14-Feb-2016
3 Comments
aaron Harvey
on 14 Feb 2016
Edited: aaron Harvey
on 14 Feb 2016
You need to reshape the string into rows of dates and convert to cell array for datenum to work:
datenum(cellstr(reshape(sprintf('%d', datevar), 8, [])', 'yyyymmdd'))
Star Strider
on 14 Feb 2016
My pleasure.
I was illustrating the idea. To use my code on a column vector of double-precision dates, it needs to be tweaked a bit:
double_date_vec = [20160214:20160313]'; % Column Vector Of Double-Precision Dates
date_number_fcn = @(double_dates) datenum(sprintf('%.0f\n', double_dates), 'yyyymmdd');
date_nums = arrayfun(date_number_fcn, double_date_vec);
Check = datestr(date_nums(1:20,:)) % Check Output (Can Be Deleted)
The date_nums array contains the datenum date numbers for all the dates in the ‘double_date_vec’ column vector. The arrayfun call does the background looping (much more efficiently than a for loop) necessary to apply the ‘date_number_fcn’ function to the vector of double-precision date variables. The MATLAB date and time functions do all the necessary conversions, including in this instance to allow for 2016 being a leap year.
I usually include ‘Check’ variables to illustrate that my code does what it needs to. Here, it prints 14-Feb-2016 to 04-Mar-2016 to the Command Window. You can discard that line.
Categories
Find more on Time Series Objects in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!